724 PROCEEDINGS OF THE AMERICAN ACADEMY. 



the hydrogen is very great compared with that of the hydride, and 

 the decomposition would not take place except at a relatively high 

 temperature. 



We have, therefore, no data as to the decomposition point of anti- 

 mony hydride in such dilution, since investigators of this point, for 

 example Brunn,^ have dealt with mixtures containing much greater 

 amounts of antimony. Brunn gives the decomposition point of anti- 

 mony hydride as 150°; of arsenic hydride as 230°; but the amount 

 of each in the mixture is not stated. Furthermore, in our work the 

 differences in the amounts of antimony are relatively so slight in com- 

 parison with the large amount of diluting hydrogen that the tempera- 

 tures required for decomposition should not vary within wide limits. 

 As the amount of antimony to be estimated in mirror form should not 

 exceed 0.1 mg, of antimonious oxide, the question to be considered is 

 the lowest temperature at which the hydride from this amount, mixed 

 with a relatively very large volume of hydrogen, would be decomposed 

 when passed through the heated tube. At the same time, with a view 

 to the possibility of separating small amounts of arsenic and antimony 

 by the difference in decomposition points of the diluted hydrides, we 

 included in our investigation the decomposition point of diluted arsenic 

 hydride. While it is well known that antimony hydride, in any dilu- 

 tion, decomposes at a lower temperature than arsenic hydride at equal 

 dilution, there were no exact data to guide us. 



At the beginning of the study, the tube was heated just behind the 

 drawn-out portion, or, as we shall call it, the capillary. It was very 

 soon found that even with small portions of antimony there was a tend- 

 ency to deposition in the wide part of the tube back of the heated 

 portion, since the larger surface offers greater chance for deposition of 

 the rapidly condensing antimony. It was then found necessary to heat 

 the capillary itself, which had the effect of concentrating the deposit 

 at the desired place just in front of the heated space. 



For determining the decomposition temperature, we used the thermo- 

 electric method of Le Chatelier, employing for the purpose a couple of 

 which the wires were platinum and an alloy of platinum with ten per 

 cent of rhodium. The couple was standardized by determining the 

 electromotive force (in micro- volts) produced by heating the junction 

 in the vapor of boiling naphthaline (218°), diphenylamine (302°), and 

 sulphur (445°). With these three boiling points as ordinates and the 

 corresponding electromotive forces as abscissae, a perfectly straight 

 line was obtained. As the temperature to be measured did not exceed 

 700° or fall below 200°, it w.s justifiable to extend this line and use 



8 Ibid., 22, 3202 (1889). 



